[tor-commits] [stem/master] Checking for mandatory and disallowed fields
atagar at torproject.org
atagar at torproject.org
Sat Oct 13 18:35:45 UTC 2012
commit d30a628e5da06699d8533f92c4b1c496510213b8
Author: Damian Johnson <atagar at torproject.org>
Date: Sun Sep 2 17:06:20 2012 -0700
Checking for mandatory and disallowed fields
There's several restrictions on a valid network status document, some of which
are which fields it does and does not contain. Validating that manitory fields
apprear, and that fields which shouldn't appear don't.
This also drops my _get_entries() helper function in favor of
_get_descriptor_components() since we're now dealing with documents which can
contain key blocks. There wasn't really enough of an advantage to
_get_entries() to justify the duplicate code anyway.
Also fixing the _get_descriptor_components() I wrote so it return footer
content in the header when the content doens't contain any 'r' lines.
This currently keeps the old parser (renamed to "_parse_old()") since I haven't
addressed any of the actual parsing yet.
---
stem/descriptor/__init__.py | 22 +++--
stem/descriptor/networkstatus.py | 154 ++++++++++++++++++++------------
test/integ/descriptor/networkstatus.py | 2 -
3 files changed, 112 insertions(+), 66 deletions(-)
diff --git a/stem/descriptor/__init__.py b/stem/descriptor/__init__.py
index fb81237..688e1b6 100644
--- a/stem/descriptor/__init__.py
+++ b/stem/descriptor/__init__.py
@@ -25,6 +25,7 @@ __all__ = [
import os
import re
import datetime
+import collections
KEYWORD_CHAR = "a-zA-Z0-9-"
WHITESPACE = " \t"
@@ -335,7 +336,7 @@ def _get_pseudo_pgp_block(remaining_contents):
else:
return None
-def _get_descriptor_components(raw_contents, validate, extra_keywords):
+def _get_descriptor_components(raw_contents, validate, extra_keywords = ()):
"""
Initial breakup of the server descriptor contents to make parsing easier.
@@ -356,13 +357,13 @@ def _get_descriptor_components(raw_contents, validate, extra_keywords):
:returns:
tuple with the following attributes...
- * **entries (dict)** - keyword => (value, pgp key) entries
+ * **entries (collections.OrderedDict)** - keyword => (value, pgp key) entries
* **first_keyword (str)** - keyword of the first line
* **last_keyword (str)** - keyword of the last line
* **extra_entries (list)** - lines containing entries matching extra_keywords
"""
- entries = {}
+ entries = collections.OrderedDict()
first_keyword = None
last_keyword = None
extra_entries = [] # entries with a keyword in extra_keywords
@@ -371,8 +372,15 @@ def _get_descriptor_components(raw_contents, validate, extra_keywords):
while remaining_lines:
line = remaining_lines.pop(0)
- # last line can be empty
- if not line and not remaining_lines: continue
+ # V2 network status documents explicitely can contain blank lines...
+ #
+ # "Implementations MAY insert blank lines for clarity between sections;
+ # these blank lines are ignored."
+ #
+ # ... and server descriptors end with an extra newline. But other documents
+ # don't say how blank lines should be handled so globally ignoring them.
+
+ if not line: continue
# Some lines have an 'opt ' for backward compatability. They should be
# ignored. This prefix is being removed in...
@@ -400,10 +408,8 @@ def _get_descriptor_components(raw_contents, validate, extra_keywords):
if keyword in extra_keywords:
extra_entries.append("%s %s" % (keyword, value))
- elif keyword in entries:
- entries[keyword].append((value, block_contents))
else:
- entries[keyword] = [(value, block_contents)]
+ entries.setdefault(keyword, []).append((value, block_contents))
return entries, first_keyword, last_keyword, extra_entries
diff --git a/stem/descriptor/networkstatus.py b/stem/descriptor/networkstatus.py
index 021ac4c..168c768 100644
--- a/stem/descriptor/networkstatus.py
+++ b/stem/descriptor/networkstatus.py
@@ -76,6 +76,34 @@ Flag = stem.util.enum.Enum(
("VALID", "Valid"),
)
+# Network status document are either a 'vote' or 'consensus', with different
+# mandatory fields for each. Both though require that their fields appear in a
+# specific order. This is an ordered listing of the following...
+#
+# (field, in_votes, in_consensus, is_mandatory)
+
+HEADER_STATUS_DOCUMENT_FIELDS = (
+ ("network-status-version", True, True, True),
+ ("vote-status", True, True, True),
+ ("consensus-methods", True, False, False),
+ ("consensus-method", False, True, False),
+ ("published", True, False, True),
+ ("valid-after", True, True, True),
+ ("fresh-until", True, True, True),
+ ("valid-until", True, True, True),
+ ("voting-delay", True, True, True),
+ ("client-versions", True, True, False),
+ ("server-versions", True, True, False),
+ ("known-flags", True, True, True),
+ ("params", True, True, False),
+)
+
+FOOTER_STATUS_DOCUMENT_FIELDS = (
+ ("directory-footer", True, True, True),
+ ("bandwidths-weights", False, True, False),
+ ("directory-signature", True, True, True),
+)
+
def parse_file(document_file, validate = True, is_microdescriptor = False):
"""
Parses a network status and iterates over the RouterStatusEntry or
@@ -94,7 +122,7 @@ def parse_file(document_file, validate = True, is_microdescriptor = False):
"""
header, footer, routers_end = _get_document_content(document_file, validate)
- document_data = "".join(header + footer)
+ document_data = header + footer
if not is_microdescriptor:
document = NetworkStatusDocument(document_data, validate)
@@ -110,7 +138,7 @@ def _get_document_content(document_file, validate):
"""
Network status documents consist of three sections: header, router entries,
and the footer. This provides back a tuple with the following...
- (header_lines, footer_lines, routers_end)
+ (header, footer, routers_end)
This leaves the document_file at the start of the router entries.
@@ -126,7 +154,7 @@ def _get_document_content(document_file, validate):
# parse until the first router record
- header = _read_until_keywords("r", document_file)
+ header = _read_until_keywords(("r", "directory-footer", "directory-signature"), document_file)
routers_start = document_file.tell()
# figure out the network status version
@@ -134,12 +162,12 @@ def _get_document_content(document_file, validate):
# TODO: we should pick either 'directory-footer' or 'directory-signature'
# based on the header's network-status-version
- _read_until_keywords(["directory-footer", "directory-signature"], document_file, skip = True)
+ _read_until_keywords(("directory-footer", "directory-signature"), document_file, skip = True)
routers_end = document_file.tell()
footer = document_file.readlines()
document_file.seek(routers_start)
- return (header, footer, routers_end)
+ return ("".join(header), "".join(footer), routers_end)
def _get_routers(document_file, validate, document, end_position, router_type):
"""
@@ -166,14 +194,11 @@ def _get_routers(document_file, validate, document, end_position, router_type):
class NetworkStatusDocument(stem.descriptor.Descriptor):
"""
- A v3 network status document.
-
- This could be a v3 consensus or vote document.
+ Version 3 network status document. This could be either a vote or consensus.
:var tuple routers: RouterStatusEntry contained in the document
- :var bool validated: **\*** whether the document is validated
- :var str network_status_version: **\*** a document format version. For v3 documents this is "3"
+ :var int network_status_version: **\*** document version
:var str vote_status: **\*** status of the vote (is either "vote" or "consensus")
:var list consensus_methods: **^** A list of supported consensus generation methods (integers)
:var datetime published: **^** time when the document was published
@@ -210,7 +235,6 @@ class NetworkStatusDocument(stem.descriptor.Descriptor):
self.directory_authorities = []
self.directory_signatures = []
- self.validated = validate
self.network_status_version = None
self.vote_status = None
@@ -231,8 +255,8 @@ class NetworkStatusDocument(stem.descriptor.Descriptor):
document_file = StringIO(raw_content)
header, footer, routers_end = _get_document_content(document_file, validate)
- document_content = "".join(header + footer)
- self._parse(document_content)
+ self._parse(header, footer, validate)
+ self._parse_old(header + footer, validate)
if document_file.tell() < routers_end:
self.routers = tuple(_get_routers(document_file, validate, self, routers_end, self._get_router_type()))
@@ -254,10 +278,30 @@ class NetworkStatusDocument(stem.descriptor.Descriptor):
return self.unrecognized_lines
- def _parse(self, raw_content):
+ def _parse(self, header, footer, validate):
+ """
+ Parses the given content and applies the attributes.
+
+ :param str header_content: content of the descriptor header
+ :param str footer_content: content of the descriptor footer
+ :param bool validate: checks validity if True
+
+ :raises: ValueError if a validity check fails
+ """
+
+ header_entries = stem.descriptor._get_descriptor_components(header, validate)[0]
+ footer_entries = stem.descriptor._get_descriptor_components(footer, validate)[0]
+
+ if validate:
+ if not 'vote-status' in header_entries:
+ raise ValueError("Network status documents must have a 'vote-status' line to say if they're a vote or consensus")
+
+ is_consensus = header_entries['vote-status'][0][0] == "consensus"
+ self._check_for_missing_and_disallowed_fields(is_consensus, header_entries, footer_entries)
+
+ def _parse_old(self, raw_content, validate):
# preamble
content = StringIO(raw_content)
- validate = self.validated
read_keyword_line = lambda keyword, optional = False: setattr(self, keyword.replace("-", "_"), _read_keyword_line(keyword, content, validate, optional))
map(read_keyword_line, ["network-status-version", "vote-status"])
@@ -320,6 +364,40 @@ class NetworkStatusDocument(stem.descriptor.Descriptor):
self.unrecognized_lines = content.read()
if validate and self.unrecognized_lines: raise ValueError("Unrecognized trailing data")
+
+ def _check_for_missing_and_disallowed_fields(self, is_consensus, header_entries, footer_entries):
+ """
+ Checks that we have mandatory fields for our type, and that we don't have
+ any fields exclusive to the other (ie, no vote-only fields appear in a
+ consensus or vice versa).
+
+ :param bool is_consensus: true if we're a conensus and false if we're a vote
+ :param dict header_entries: ordered keyword/value mappings of the header
+ :param dict footer_entries: ordered keyword/value mappings of the footer
+
+ :raises: ValueError if we're missing mandatory fields or have fiels we shouldn't
+ """
+
+ is_vote = not is_consensus # aliasing inverse for readability
+ missing_fields, disallowed_fields = [], []
+
+ for entries, fields in ((header_entries, HEADER_STATUS_DOCUMENT_FIELDS),\
+ (footer_entries, FOOTER_STATUS_DOCUMENT_FIELDS)):
+ for field, in_votes, in_consensus, mandatory in fields:
+ if mandatory and ((is_consensus and in_consensus) or (is_vote and in_votes)):
+ # mandatory field, check that we have it
+ if not field in entries.keys():
+ missing_fields.append(field)
+ elif (is_consensus and not in_consensus) or (is_vote and not in_votes):
+ # field we shouldn't have, check that we don't
+ if field in entries.keys():
+ disallowed_fields.append(field)
+
+ if missing_fields:
+ raise ValueError("Network status document is missing mandatory field: %s" % ', '.join(missing_fields))
+
+ if disallowed_fields:
+ raise ValueError("Network status document has fields that shouldn't appear in this document type: %s" % ', '.join(disallowed_fields))
class DirectoryAuthority(stem.descriptor.Descriptor):
"""
@@ -509,7 +587,10 @@ class RouterStatusEntry(stem.descriptor.Descriptor):
:raises: ValueError if a validity check fails
"""
- entries = _get_entries(content, validate, 'r')
+ entries, first_keyword, _, _ = stem.descriptor._get_descriptor_components(content, validate)
+
+ if validate and first_keyword != 'r':
+ raise ValueError("Router status entries are expected to start with a 'r' line:\n%s" % (content))
# check that we have mandatory fields
if validate:
@@ -518,7 +599,7 @@ class RouterStatusEntry(stem.descriptor.Descriptor):
raise ValueError("Router status entries must have a '%s' line:\n%s" % (keyword, content))
for keyword, values in entries.items():
- value = values[0]
+ value, block_contents = values[0]
line = "%s %s" % (keyword, value)
# most attributes can only appear at most once
@@ -685,7 +766,6 @@ class MicrodescriptorConsensus(NetworkStatusDocument):
"""
A v3 microdescriptor consensus.
- :var bool validated: **\*** whether the document is validated
:var str network_status_version: **\*** a document format version. For v3 microdescriptor consensuses this is "3 microdesc"
:var str vote_status: **\*** status of the vote (is "consensus")
:var int consensus_method: **~** consensus method used to generate a consensus
@@ -838,44 +918,6 @@ class RouterMicrodescriptor(RouterStatusEntry):
return self.unrecognized_lines
-def _get_entries(content, validate, expected_first_keyword = None):
- """
- Provides the {keyword => [values...]} mappings for the given content.
-
- :param str content: descriptor content
- :param bool validate: checks validity if True
- :param str expected_first_keyword: validates that this is the first keyword
-
- :returns: dict with the mapping of keywords to their values
-
- :raises: ValueError if a validity check fails
- """
-
- entries = {}
-
- for line in content.split("\n"):
- # empty lines are allowed
- if not line: continue
-
- line_match = stem.descriptor.KEYWORD_LINE.match(line)
-
- if not line_match:
- if not validate: continue
- raise ValueError("Line contains invalid characters: %s" % line)
-
- keyword, value = line_match.groups()
- if value is None: value = ''
-
- if expected_first_keyword != None:
- if validate and expected_first_keyword != keyword:
- raise ValueError("Expected to start with a '%s' line:\n%s" % (expected_first_keyword, content))
-
- expected_first_keyword = None
-
- entries.setdefault(keyword, []).append(value)
-
- return entries
-
def _decode_fingerprint(identity, validate):
"""
Decodes the 'identity' value found in consensuses into the more common hex
diff --git a/test/integ/descriptor/networkstatus.py b/test/integ/descriptor/networkstatus.py
index c21b4a2..4ca8e5c 100644
--- a/test/integ/descriptor/networkstatus.py
+++ b/test/integ/descriptor/networkstatus.py
@@ -85,7 +85,6 @@ class TestNetworkStatus(unittest.TestCase):
router1 = desc.routers[0]
descriptor_file.close()
- self.assertEquals(True, desc.validated)
self.assertEquals("3", desc.network_status_version)
self.assertEquals("consensus", desc.vote_status)
self.assertEquals([], desc.consensus_methods)
@@ -178,7 +177,6 @@ I/TJmV928na7RLZe2mGHCAW3VQOvV+QkCfj05VZ8CsY=
router1 = desc.routers[0]
descriptor_file.close()
- self.assertEquals(True, desc.validated)
self.assertEquals("3", desc.network_status_version)
self.assertEquals("vote", desc.vote_status)
self.assertEquals(range(1, 13), desc.consensus_methods)
More information about the tor-commits
mailing list